Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -5629,7 +5629,10 @@ export async function recordDispatch(
-- the salvage committed already on the branch. That hand-back is
-- counted here and bounded by MAX_CEILING_RELEASES, so the swarm
-- cannot spend an issue's attempts twice over. send_backs starts
-- again with it: the new attempt is judged on its own work.
-- again at 0, but the hand-back is ONE attempt, not a new set:
-- with ceiling_releases now at MAX_RELEASES, its first send-back
-- reaches the review valve, which closes the issue as obsolete
-- (stateOfDispatch in queen-tick.ts; gHashTag/trios#1769).
-- An escalation the review valve released is the same hand-back
-- and spends the same count (specs/queen/review_valve.t27,
-- MAX_RELEASES), so the two roads cannot each spend it.
Expand Down
13 changes: 9 additions & 4 deletions trios/agent-server/apps/server/src/api/services/queen-tick.ts
Original file line number Diff line number Diff line change
Expand Up @@ -781,10 +781,15 @@ export const CEILING_RELEASE_MS = CEILING_FLOOR_MINUTES * 60 * 1000
/**
* How many times a spent ceiling may be handed back to the swarm.
*
* One. The bee gets `maximumSendBacks` attempts, then an hour, then one more
* set - with the last review's findings in its brief and whatever the salvage
* committed already on the branch. After that the row stays rejected: a third
* identical failure is evidence about the issue, not about the attempt.
* One. The bee gets `maximumSendBacks` attempts, then an hour, then ONE more
* attempt - not one more set - with the last review's findings in its brief
* and whatever the salvage committed already on the branch. The redispatch
* resets `send_backs` but carries `ceiling_releases` = 1, so its first
* send-back goes to the review valve with `releases` at `MAX_RELEASES` and
* `next_step` answers STEP_CLOSE: the claim is `cancelled` in the next round
* and the row is closed as obsolete, inside the send-back's hour
* (gHashTag/trios#1769). A third identical failure is evidence about the
* issue, not about the attempt.
*/
export const MAX_CEILING_RELEASES = MAX_RELEASES

Expand Down
Loading